Bee Venom-Loaded Niosomes for Breast Cancer: In Silico Target Prediction, Formulation, Characterization and in Vitro Cytotoxicity Against MCF-7 Cells

Honeybee venom is a mixture of peptides, enzymes and small amines, of which melittin, phospholipase A2 and apamin are the best studied.^1-3 Melittin makes up the largest share of the dry venom and is a cationic, amphipathic peptide that inserts readily into lipid bilayers.⁴ Bee venom and melittin have slowed the growth of several cancer cell types and triggered apoptosis in them, and reports in breast cancer models include MCF-7 cells and HER2-enriched and triple-negative disease.^5-9

The membrane activity that makes melittin cytotoxic to tumour cells is not selective. Free melittin and whole venom also damage normal cells, so the margin between an effective and a harmful dose is narrow, and a peptide is further limited by poor stability and uncontrolled release.^4,9 Encapsulating the venom in a carrier is one way to address these problems.

Niosomes are vesicles assembled from non-ionic surfactants and cholesterol. Their size, entrapment and release behaviour can be adjusted through the surfactant type, the surfactant-to-cholesterol ratio and the preparation conditions.^10,11 Melittin-loaded niosomes have been tested against breast cancer,¹² whole bee venom has been loaded into niosomes and into nanoliposomes,^13,14 and the venom nanoformulations were cytotoxic to cancer cell lines in those studies. What is still lacking is a stepwise assessment that links computational prediction of the likely targets to the formulation and its cell-level activity.

We therefore combined three steps in this work. First, melittin, the principal peptide of the venom, was examined by network pharmacology, molecular docking and ADMET prediction to identify breast cancer-related targets. Second, whole bee venom was formulated into Span 60/Tween 60/cholesterol niosomes and characterised for compatibility, size, surface charge, entrapment, morphology and release. Third, the formulation was tested for cytotoxicity against MCF-7 breast cancer cells.

MATERIALS AND METHODS

MATERIALS

Whole bee venom was purchased from One Bee Organic LLC. Span 60, Tween 60, cholesterol, chloroform and phosphate-buffered saline (PBS, pH 7.4) were used for vesicle preparation, and the remaining reagents were of analytical grade. MCF-7 human breast cancer cells were used for the cytotoxicity study.

In silico analysis

Melittin was chosen for the computational work because it is the main active peptide of the venom and its structure is available. Its structure was taken from PubChem¹⁵ and its potential targets were predicted with SuperPred. Genes associated with breast cancer were collected from GeneCards using the search term “breast cancer”.¹⁶ The two lists were intersected in Venny 2.1.0 to obtain the shared targets. These were submitted to STRING¹⁷ and the resulting protein–protein interaction network was opened in Cytoscape,¹⁸ where the CytoHubba plugin ranked the nodes and the ten top-ranked hub targets were retained.

Crystal structures of the ten hub proteins were downloaded from the Protein Data Bank. Melittin was docked to each protein separately on the HDOCK web server.¹⁹ The docking scores were compared across targets and the best-scoring complexes were inspected visually. Absorption, distribution, metabolism, excretion and toxicity properties of melittin were predicted with the pkCSM server.²⁰

Preparation of niosomes

Niosomes were made by thin-film hydration.^10,11 Span 60, Tween 60 and 1:1:2 (w/w/w) were dissolved in chloroform in a round-bottom flask, and the solvent was evaporated under reduced pressure in a rotary evaporator to leave a thin film. Bee venom, 5 mg, was dissolved in PBS (pH 7.4) and used to hydrate the film with magnetic stirring for 60 minutes. The dispersion was then sonicated with a probe sonicator for about 15 mintues to reduce vesicle size. Blank niosomes were prepared in the same way without venom.

Preformulation studies

The solubility of bee venom in distilled water and PBS (pH 7.4) was judged by visual inspection after mixing and equilibration. The absorption maximum was found by scanning a venom solution in a UV-visible spectrophotometer, and a calibration curve was drawn from solutions of 20–100 µg/mL read at that wavelength. For the compatibility study, FTIR spectra of bee venom, Span 60, Tween 60 and cholesterol were recorded individually and for their physical mixture, and the position of the characteristic bands was compared.

Characterisation of the niosomes

Mean vesicle size and polydispersity index (PDI) were measured by dynamic light scattering, and zeta potential by electrophoretic light scattering, on suitably diluted dispersion. To determine entrapment efficiency, unentrapped venom was separated from the vesicles by centrifugation at 2 - 4°c, 14000 rpm and measured at the absorption maximum by using UV-Visible spectrophometry. Entrapment efficiency (%) was calculated as (total venom − unentrapped venom)/total venom × 100. Vesicle morphology was examined by transmission electron microscopy.

Release was studied by the dialysis-bag method. Pure venom and the niosomal dispersion were placed in separate dialysis membranes immersed in the release medium (PBS, Ph 7.4) and 1ml of samples were withdrawn and replaced with fresh medium at 0.5, 1, 2, 4, 6, 8, 12 and 24 h. The venom released was read spectrophotometrically, and the cumulative release was fitted to zero-order, first-order, Higuchi and Korsmeyer–Peppas models.

Cytotoxicity assay

MCF-7 cells were grown in a humidified CO2 incubator, seeded into 96-well plates and allowed to attach. They were exposed for 48 h to the niosomal formulation at 10, 50, 100, 200 and 400 µg/mL, with untreated cells as the control and camptothecin (7.8 µg/mL) as the reference drug. Viability was determined by the MTT assay;²¹ formazan crystals were dissolved and absorbance was read at 570 nm. Viability (%) was calculated as the absorbance of treated cells divided by that of control cells × 100, and cytotoxicity (%) as 100 minus viability. The IC50 was obtained from the logarithmic regression of viability on concentration.

RESULTS AND DISCUSSION

Network pharmacology and target selection

Melittin targets predicted by SuperPred overlapped with the breast cancer gene set, giving a group of shared proteins (Fig. 1). The protein–protein interaction network built from these showed a densely connected core, and ranking by CytoHubba selected ten hub targets for docking (Fig. 2 and 3). The overlap suggests that melittin may act on several nodes of the disease network at once, a pattern in line with its reported effects on proliferation, apoptosis, invasion and angiogenesis.⁹ These are predicted associations only and have not been shown experimentally.

` Figure 1: Venn Diagram of Targets Shared Between Melittin and Breast Cancer.

Figure 2: Protein–Protein Interaction Network of the Shared Targets.

Figure 3: Top Ten Hub Targets Identified by CytoHubba.

Molecular docking

Melittin gave negative HDOCK scores with all ten hub proteins (Table 1). The two most favourable were MAPK1 (−233.01) and PIK3R1 (−232.85), followed by NFKB1 (−216.70) and STAT3 (−216.14); ESR1 gave the weakest score (−197.04). The poses of melittin with MAPK1 and PIK3R1 are shown in Fig. 4 and 5. MAPK1 belongs to the MAPK cascade and PIK3R1 encodes a regulatory subunit of PI3K, and both pathways control cell growth and survival. Melittin has been reported to suppress PI3K/Akt/mTOR signalling in breast cancer cells²² and PI3K/AKT/mTOR and MAPK signalling in melanoma cells,²³ which is consistent with these docking results. A favourable docking score indicates a predicted fit, not inhibition of the protein, so these targets need experimental confirmation.

Table 1: Docking Scores of Melittin with the Ten HUB Targets

Figure 4: Predicted Interaction of Melittin with MAPK1 (PDB ID 2OJJ).

Figure 5: Predicted Interaction of Melittin with PIK3R1 (PDB ID 4OVU).

ADMET prediction

The pkCSM output for melittin is summarised in Table 2. Intestinal absorption was predicted as 0%, which is expected for a peptide and supports a non-oral delivery strategy and a carrier. Melittin was predicted to be a P-glycoprotein substrate and a CYP3A4 substrate, with no inhibition of the five CYP isoforms tested, no AMES toxicity and no hERG I inhibition. Hepatotoxicity and hERG II inhibition were, however, predicted. Because pkCSM was developed on small molecules, its values for a 26-residue peptide should be read as indicative, and the predicted toxicity flags are a reason for experimental safety testing rather than evidence of harm. The computation also used melittin alone, so it speaks to the likely contribution of this peptide and not to the whole venom.

Table 2: Selected Predicted Admet Properties of Melittin (pkCSM)

Preformulation: solubility, UV method and FTIR

Bee venom dissolved in both distilled water and PBS (pH 7.4), and PBS was chosen as the hydration medium. The venom had an absorption maximum at 278 nm (Fig. 6). Absorbance was linear over 20–100 µg/mL, with the regression y = 0.00517x − 0.039 and R² = 0.9983 (Fig. 7), which was adequate for estimating venom in the entrapment and release experiments. In the FTIR study, the characteristic amide, ester and alkyl bands of the individual components were present in the physical mixture without notable shifts or loss of the main venom peaks (Fig. 8), so no evident interaction between venom and excipients was seen.

Figure 6: UV-visible Absorption Spectrum of Bee Venom (Λmax 278 nm).

Figure 7: Calibration Curve of Bee Venom, 20–100 Μg/ml.

Figure 8: FTIR Spectra of Bee Venom, Span 60, Tween 60, Cholesterol and Their Physical Mixture.

Physicochemical characteristics of the niosomes

The dispersion was milky in appearance, as expected for vesicle formation. The optimised formulation had a mean size of 208.7 nm and a PDI of 0.3219, which indicates a moderately broad distribution within the nanometre range (Fig. 9). The zeta potential was −21.51 mV (Fig. 10); a negative surface charge of this size can provide some electrostatic repulsion between vesicles. The unentrapped venom gave an absorbance of 0.209, corresponding to 47.72 µg/mL, or 0.764 mg in 16 mL, out of 5 mg added, so the entrapment efficiency was 84.73% (Table 3). TEM showed rounded to oval vesicles with distinct boundaries, some variation in size and shape, and clustering in places (Fig. 11). The sizes seen by TEM were smaller than the hydrodynamic diameter from DLS, which is commonly seen because DLS measures the hydrated vesicle in suspension and TEM the dried sample.

Table 3: Physicochemical Characteristics of Bee Venom-Loaded Niosomes

Figure 9: Particle Size Distribution and Polydispersity Index (PDI) of Bee Venom-Loaded Niosomes.

`

Figure 10: Zeta Potential Distribution of Bee Venom-Loaded Niosomes.

Figure 11: TEM Images of Bee Venom-Loaded Niosomes.

In vitro release

Pure venom released 16.70% at 0.5 h, 71.32% at 6 h and 93.92% at 24 h, whereas the niosomes released 8.43%, 28.17% and 89.18% at the same times (Table 4). Release from the niosomes was therefore much slower over the first 6 h, which is the clearest effect of encapsulation, but it rose sharply between 6 and 8 h (28.17% to 63.36%) and by 12 h was close to that of the pure venom (82.33% against 83.99%). The vesicle bilayer thus delayed early release but did not sustain it over the full 24 h. A different surfactant-to-cholesterol ratio could be explored if a longer plateau is required.

Table 4: Cumulative in Vitro Release of Bee Venom from Pure Venom and Niosomes

Of the four models, Korsmeyer–Peppas gave the highest coefficient of determination for both pure venom (R² = 0.936) and niosomes (R² = 0.984), and the Higuchi model came next (0.876 and 0.885) (Table 5, Fig. 13). Diffusion therefore appears to contribute to release, suggesting that the release mechanism may involve both diffusion and structural relaxation, as indicated by the Korsmeyer-Peppas model.

Table 5: Coefficients of Determination (R²) for Release Kinetic Models

[Figure - see the formatted PDF]

[Figure - see the formatted PDF]

[Figure - see the formatted PDF]

[Figure - see the formatted PDF]

Figure 13: Release Kinetic Plots: (A) Zero-Order, (B) First-Order, (C) Higuchi and (D) Korsmeyer–Peppas.

Cytotoxicity against MCF-7 cells

After 48 h the formulation lowered MCF-7 viability in a concentration-dependent way (Table 6, Fig. 14). Viability was 91.00% at 10 µg/mL and 84.20% at 50 µg/mL, fell to 58.32% at 100 µg/mL and 49.74% at 200 µg/mL, and reached 21.31% at 400 µg/mL. The IC50 from the logarithmic fit (y = −18.47 ln x + 142.59, R² = 0.8689) was 150.35 µg/mL (Fig. 15). The reference drug camptothecin at 7.8 µg/mL left 59.02% of cells viable. The MTT assay reports on metabolic activity and cannot by itself show how the cells died, and the study did not include free bee venom or blank niosomes as comparators, so the contribution of the carrier and any gain over the free venom remain to be established. The IC50 here is also not directly comparable with values from other niosomal venom studies that used different assays and exposure times.^13,24

Table 6: Effect of Bee Venom-Loaded Niosomes on MCF-7 Cell Viability at 48 H

Figure 14: Viability of MCF-7 Cells After 48 H Exposure to Bee Venom-Loaded Niosomes.

[Figure - see the formatted PDF]

Figure 15: Determination of the IC50 of Bee Venom-Loaded Niosomes Against MCF-7 Cells.

CONCLUSION

Computational analysis pointed to MAPK1 and PIK3R1, among ten hub targets, as the best-scoring partners of melittin and flagged poor oral absorption and possible hepatotoxicity, which supports a delivery system and further safety work. Bee venom was successfully formulated into Span 60/Tween 60/cholesterol niosomes of about 209 nm with an entrapment efficiency of 84.73%, and no evident incompatibility with the excipients. The vesicles slowed the early release of venom and reduced MCF-7 viability in a concentration-dependent manner, with an IC50 of 150.35 µg/mL. These results justify testing the formulation in vivo; comparison with free venom and blank vesicles and mechanistic studies are needed to define the benefit of the carrier.

ACKNOWLEDGEMENT

The authors thank the Faculty of Pharmacy, M. S. Ramaiah University of Applied Sciences, Bengaluru, for providing the facilities needed for this work.

FUNDING

This research received no external funding.

CONFLICT OF INTEREST

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

REFERENCES

1. Habermann E: Bee and wasp venoms. Science 1972; 177 (4046):314-322.
2. Hider RC: Honeybee venom: a rich source of pharmacologically active peptides. Endeavour 1988; 12 (2):60-65.
3. Son DJ, Lee JW, Lee YH, Song HS, Lee CK and Hong JT: Therapeutic application of anti-arthritis, pain-releasing, and anti-cancer effects of bee venom and its constituent compounds. Pharmacology & Therapeutics 2007; 115 (2):246-270.
4. Raghuraman H and Chattopadhyay A: Melittin: a membrane-active peptide with diverse functions. Bioscience Reports 2007; 27 (4-5):189-223.
5. Ip SW, Liao SS, Lin SY, Lin JP, Yang JS, Lin ML, Chen GW, Lu HF, Lin MW, Han SM and Chung JG: The role of mitochondria in bee venom-induced apoptosis in human breast cancer MCF7 cells. In Vivo 2008; 22 (2):237-245.
6. Yeo SW, Seo JC, Choi YH and Jang KJ: Induction of the growth inhibition and apoptosis by bee venom in human breast carcinoma MCF-7 cells. Journal of Korean Acupuncture and Moxibustion Medicine Society 2003; 20:45-62.
7. Oršolić N, Šver L, Verstovšek S, Terzić S and Bašić I: Inhibition of mammary carcinoma cell proliferation in vitro and tumor growth in vivo by bee venom. Toxicon 2003; 41 (7):861-870.
8. Duffy C, Sorolla A, Wang E, Golden E, Woodward E, Davern K, Ho D, Johnstone E, Pfleger K, Redfern A, Iyer KS, Baer B and Blancafort P: Honeybee venom and melittin suppress growth factor receptor activation in HER2-enriched and triple-negative breast cancer. npj Precision Oncology 2020; 4:24.
9. Kwon NY, Sung SH, Sung HK and Park JK: Anticancer activity of bee venom components against breast cancer. Toxins 2022; 14 (7):460.
10. Baillie AJ, Florence AT, Hume LR, Muirhead GT and Rogerson A: The preparation and properties of niosomes - non-ionic surfactant vesicles. Journal of Pharmacy and Pharmacology 1985; 37 (12):863-868.
11. Moghassemi S and Hadjizadeh A: Nano-niosomes as nanoscale drug delivery systems: an illustrated review. Journal of Controlled Release 2014; 185:22-36.
12. Dabbagh Moghaddam F, Akbarzadeh I, Marzbankia E, Farid M, Khaledi L, Reihani AH, Javidfar M and Mortazavi P: Delivery of melittin-loaded niosomes for breast cancer treatment: an in vitro and in vivo evaluation of anti-cancer effect. Cancer Nanotechnology 2021; 12:14.
13. Pinto MB, Pires PC, Calhelha RC, Silva AR, Sousa MJ, Vilas-Boas M, Falcão SI, Veiga F, Makvandi P and Paiva-Santos AC: Bee venom-loaded niosomes as innovative platforms for cancer treatment: development and therapeutical efficacy and safety evaluation. Pharmaceuticals 2024; 17 (5):572.
14. Abd El-Gawad A, Kenawy MA, El-Messery TM, Hassan ME, El-Nekeety AA and Abdel-Wahhab MA: Fabrication and characterization of bee venom-loaded nanoliposomes: enhanced anticancer activity against different human cancer cell lines via the modulation of apoptosis-related genes. Journal of Drug Delivery Science and Technology 2023; 84:104545.
15. Kim S, Thiessen PA, Bolton EE, Chen J, Fu G, Gindulyte A et al.: PubChem Substance and Compound databases. Nucleic Acids Research 2016; 44 (D1):D1202-D1213.
16. Stelzer G, Rosen N, Plaschkes I, Zimmerman S, Twik M, Fishilevich S et al.: The GeneCards Suite: from gene data mining to disease genome sequence analyses. Current Protocols in Bioinformatics 2016; 54:1.30.1-1.30.33.
17. Szklarczyk D, Kirsch R, Koutrouli M, Nastou K, Mehryary F, Hachilif R et al.: The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Research 2023; 51 (D1):D638-D646.
18. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B and Ideker T: Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Research 2003; 13 (11):2498-2504.
19. Yan Y, Zhang D, Zhou P, Li B and Huang SY: HDOCK: a web server for protein-protein and protein-DNA/RNA docking based on a hybrid strategy. Nucleic Acids Research 2017; 45 (W1):W365-W373.
20. Pires DEV, Blundell TL and Ascher DB: pkCSM: predicting small-molecule pharmacokinetic and toxicity properties using graph-based signatures. Journal of Medicinal Chemistry 2015; 58:4066-4072.
21. Mosmann T: Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. Journal of Immunological Methods 1983; 65 (1-2):55-63.
22. Jeong YJ, Choi Y, Shin JM, Cho HJ, Kang JH, Park KK et al.: Melittin suppresses EGF-induced cell motility and invasion by inhibiting PI3K/Akt/mTOR signaling pathway in breast cancer cells. Food and Chemical Toxicology 2014; 68:218-225.
23. Lim HN, Baek SB and Jung HJ: Bee venom and its peptide component melittin suppress growth and migration of melanoma cells via inhibition of PI3K/AKT/mTOR and MAPK pathways. Molecules 2019; 24 (5):929.
24. Lischer K, Sitorus SRA, Guslianto BW, Avila F, Khayrani AC and Sahlan M: Anti-breast cancer activity on MCF-7 cells of melittin from Indonesia's Apis cerana: an in vitro study. Asian Pacific Journal of Cancer Prevention 2021; 22:3913-3919.